Compare / Qwen3.5-9B vs Claude Haiku 4.5
Qwen3.5-9B vs Claude Haiku 4.5
Which is cheaper, which is more capable, and which does better at each kind of work, from 15 results on 2 sources that measured both models.
Of the 15 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee). See also Alibaba vs Anthropic, every model from both labs.
Qwen3.5-9B or Claude Haiku 4.5? The short answer
Pick Claude Haiku 4.5 if quality matters most: it scores higher (Intelligence Index 16.9 against 13.3). Pick Qwen3.5-9B if cost matters more: it costs 18 times less per token.
- Qwen3.5-9B is cheaper: $0.11 against $2.00 per million tokens, 18 times less.
- Claude Haiku 4.5 scores higher on the Artificial Analysis Intelligence Index: 16.9 against 13.3.
- Qwen3.5-9B is newer: released 2 Mar 2026, against 15 Oct 2025 for Claude Haiku 4.5.
- On third-party benchmarks, on the results both models have: Qwen3.5-9B does better for reasoning and knowledge; Claude Haiku 4.5 for agents and tool use and writing.
Head to head
Every fact side by side
| Qwen3.5-9B | Claude Haiku 4.5 | |
|---|---|---|
| Price per million tokens | $0.10 in · $0.15 out | $1.00 in · $5.00 out |
| Blended (3 in : 1 out) | $0.11 | $2.00 |
| Cheapest host, blended | $0.092 (Darkbloom, FP4) | $2.00 (Amazon Bedrock) |
| Cached input, per million | – | $0.10 cached input |
| Intelligence Index | 13.3 | 16.9 |
| Spring Prompt overall | – | 8 |
| Output speed, tokens a second | 110 | – |
| Context window | 262,144 tokens | 200,000 tokens |
| Released | 2 Mar 2026 (newer) | 15 Oct 2025 |
| Weights | Open | Closed |
| Developer based in | China | the US |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.
From our research
Every shared result
Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.
| Measure | Qwen3.5-9B | Claude Haiku 4.5 |
|---|---|---|
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 145 · 122 of 249 measured | 13.3no reasoning | 16.9thinking reasoning |
| GPQA Diamond% of questions, higher is better 109 · 164 of 232 measured | 80.6% | 67.2%thinking reasoning |
| Humanity's Last Exam% of questions, higher is better 134 · 157 of 250 measured | 14.9% | 10.4%thinking reasoning |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 113 · 72 of 230 measured | 39.5think-prefill reasoning | 52.2no reasoning |
| Requested-length error · UGI% off the requested word count, lower is better 69 · 17 of 230 measured | 15.0%think-prefill reasoning | 6.0%no reasoning |
| Style adherence · UGIscore from 0 to 1, higher is better 173 · 22 of 230 measured | 0.32no reasoning | 0.39thinking reasoning |
Show all 15 results (9 more)
| Measure | Qwen3.5-9B | Claude Haiku 4.5 |
|---|---|---|
| Artificial Analysis reported by Artificial Analysis · 9 results | ||
| Terminal-Bench 2.1% of tasks, higher is better 84 · 71 of 131 measured | 29.2% | 44.2%thinking reasoning |
| Terminal-Bench 4.0% of tasks, higher is better 65 · 72 of 120 measured | 0.5% | 0.0%thinking reasoning |
| τ-bench banking% of tasks, higher is better 90 · 82 of 121 measured | 7.0% | 9.3%thinking reasoning |
| Artificial Analysis Coding Indexindex score, higher is better 87 · 67 of 130 measured | 28.7 | 43.9thinking reasoning |
| IFBench% of instructions, higher is better 68 · 92 of 195 measured | 66.7% | 54.3%thinking reasoning |
| Long-context reasoning (AA-LCR)% of questions, higher is better 116 · 89 of 241 measured | 70.0% | 74.3%thinking reasoning |
| SciCode% of problems, higher is better 110 · 81 of 121 measured | 29.5% | 42.2%thinking reasoning |
| Terminal-Bench Hard% of tasks, higher is better 80 · 72 of 190 measured | 24.2% | 27.3%no reasoning |
| τ²-bench telecom% of tasks, higher is better 51 · 98 of 193 measured | 86.8% | 54.7%thinking reasoning |
Compare them with other models
Quick answers
Is Qwen3.5-9B better than Claude Haiku 4.5?
Pick Claude Haiku 4.5 if quality matters most: it scores higher (Intelligence Index 16.9 against 13.3). Pick Qwen3.5-9B if cost matters more: it costs 18 times less per token.
Which is cheaper, Qwen3.5-9B or Claude Haiku 4.5?
Qwen3.5-9B is cheaper: $0.11 against $2.00 per million tokens, 18 times less.
Which is better for agents and tool use, Qwen3.5-9B or Claude Haiku 4.5?
Claude Haiku 4.5 ranks 98 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 111 for Qwen3.5-9B.
Which is better for reasoning and knowledge, Qwen3.5-9B or Claude Haiku 4.5?
Qwen3.5-9B ranks 116 of 187 models on sale for hard questions across science, maths and general knowledge, against 124 for Claude Haiku 4.5.
Which is better for writing, Qwen3.5-9B or Claude Haiku 4.5?
Claude Haiku 4.5 ranks 87 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 112 for Qwen3.5-9B.
Which has the bigger context window, Qwen3.5-9B or Claude Haiku 4.5?
Qwen3.5-9B: 262,144 tokens, against 200,000 for Claude Haiku 4.5.
Run Qwen3.5-9B and Claude Haiku 4.5 on your own prompt.
Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.